Visual malfunction is a common corollary of stroke and several other types of cortical injury. Posterior circulation infarcts, hemorrhages or traumati brain injury often produce varying degrees of damage to visual cortical networks, including the primary visual cortex, resulting in partial or complete homonymous hemianopias or quadrantanopias. The most common clinically significant visual cortical injury involves the primary visual cortex (V1). V1 is the chief relay of visual input to higher (extra striate) cortica areas and V1 lesions result in a dense contralateral scotoma within which visual perception is severely impaired. The resulting visual deficit is long thought to be highly resistant to rehabilitation, i.e. essentially irreversible. Hope, however, may not be entirely lost. Under carefully controlled conditions a limited capacity to process visual attributes such as motion often persists inside scotomas induced by V1 lesions, both in humans and monkeys (blindsight). Blindsight performance improves with training (Weiskrantz L, Prog Brain Res, 144:229-41, 2004; Huxlin et al., J Neurosci. 29(13):3981-91, 2009) raising the hope that better rehabilitation strategies may one day be able to increase the strength of V1-bypassing pathways to partially compensate for the loss of V1 input. Visual rehabilitation however remains challenging and the capacity of individual patients for visual rehabilitation highly variable (fig.3). It is important to understand which areas of the visual field are more amenable to rehabilitation and to study possible mechanisms underlying recovery. Functional magnetic resonance imaging (fMRI) can be used to map population receptive field (pRF) properties in normal visual cortex (Dumoulin et al., J Neurosci, 2008). This offers a unique opportunity to characterize in detail, voxel by voxel, how pRFs in spared visual areas are organized following area V1+ injuries. However, in order to do so, the pRF mapping methodology has to be refined to eliminate biases that occur in pRF estimation near the border of a perceptual scotoma (fig. 1, 5, 6). We will study a patient cohort with chronic hemianopia and quadrantanopia as a result of V1+ lesions and age-matched controls. Specific aim #1 will develop a direct method of estimating pRF topography that is less prone to bias (figs 1, 5, 6). Specific aim #2 will characterize how visual field representation and pRF properties in spared visual areas differ in subjects with chronic V1+ lesions compared to controls. The capacity of spared visual areas to be modulated from visual field locations within the perceptual scotoma will identify locations more amenable to rehabilitation. Specific aim #3 will test the hypothesis that rehabilitative training in visual moton perception improves performance by increasing the sensitivity of the motion selective complex (hV5/MT+) to visual motion stimuli. This effect will be particularly strong in regions of the scotoma that can visually modulate area hV5/MT+ before training (SA #2). Overall, our approach will characterize how downstream areas adjust to chronic V1+ injury, and will suggest new ways to guide rehabilitative training in the future.